The neuropeptide galanin and its receptors hold great promise as targets for the development of novel therapeutic agents. Galanin is widely distributed throughout the peripheral and central nervous systems and is associated with the regulation of processes such as somatosensory transmission, smooth muscle contractility, hormone release, and feeding (for review, see Bartfai et al., 1993). In the periphery galanin is found in the adrenal medulla, uterus, gastrointestinal tract, dorsal root ganglia (DRG), and sympathetic neurons. Galanin released from sympathetic nerve terminals in the pancreas is a potent regulator of insulin release in several species (Ahren and Lindskog, 1992; Boyle et al., 1994), suggesting a potential role for galanin in the etiology or treatment of diabetes. High levels of galanin are observed in human and rat anterior pituitary where galanin mRNA levels are potently upregulated by estrogen (Vrontakis et al., 1987; Kaplan et al., 1988). The presence of galanin in the hypothalamic-pituitary-adrenal axis coupled with its potent hormonal effects has led to the suggestion that galanin may play an integral role in the hormonal response to stress (Bartfai et al., 1993).
Within the CNS galanin-containing cell bodies are found in the hypothalamus, hippocampus, amygdala, basal forebrain, brainstem nuclei, and spinal cord, with highest concentrations of galanin in the hypothalamus and pituitary (Skofitsch and Jacobowitz, 1985; Bennet et al., 1991; Merchenthaler et al., 1993). The distribution of galanin receptors in the CNS generally complements that of galanin peptide, with high levels of galanin binding observed in the hypothalamus, amygdala, hippocampus, brainstem and dorsal spinal cord (Skofitsch et al., 1986; Merchenthaler et al., 1993; see Bartfai et al., 1993). Accordingly, agents modulating the activity of galanin receptors would have multiple potential therapeutic applications in the CNS. One of the most important of these is the regulation of food intake. Galanin injected into the paraventricular nucleus (PVN) of the hypothalamus stimulates feeding in satiated rats (Kyrkouli et al., 1990), an effect which is blocked by the peptide galanin antagonist M40 (Crawley et al., 1993). In freely feeding rats, PVN injection of galanin preferentially stimulates fat-preferring feeding (Tempel et al., 1988); importantly, the galanin antagonist M40 administered alone decreases overall fat intake (Leibowitz and Kim, 1992). These data indicate that specific receptors in the hypothalamus mediate the effects of galanin on feeding behavior, and further suggest that agents acting at hypothalamic galanin receptors may be therapeutically useful in the treatment of human eating disorders.
Galanin receptors elsewhere in the CNS may also serve as therapeutic targets. In the spinal cord galanin is released from the terminals of sensory neurons as well as spinal interneurons and appears to play a role in the regulation of pain threshold (Wiesenfeld-Hallin et al., 1992). Intrathecal galanin potentiates the anti-nociceptive effects of morphine in rats and produces analgesia when administered alone (Wiesenfeld-Hallin et al., 1993; Post et al., 1988); galanin receptor agonists may therefore be useful as analgesic agents in the spinal cord. Galanin may also play a role in the development of Alzheimer's disease. In the hippocampus galanin inhibits both the release (Fisone et al., 1987) and efficacy (Palazzi et al., 1988) of acetylcholine, causing an impairment of cognitive functions (Sundstrom et al., 1988). Autopsy samples from humans afflicted with Alzheimer's disease reveal a galaninergic hyperinnervation of the nucleus basalis (Chan-Palay, 1988), suggesting a role for galanin in the impaired cognition characterizing Alzheimer's disease. Together these data suggest that a galanin antagonist may be effective in ameliorating the symptoms of Alzheimer's disease (see Crawley, 1993). This hypothesis is supported by the report that intraventricular administration of the peptide galanin antagonist M35 improves cognitive performance in rats (Ogren et al., 1992). Human galanin receptors thus provide targets for therapeutic intervention in multiple CNS disorders.
High-affinity galanin binding sites have been characterized in brain, spinal cord, pancreatic islets and cell lines, and gastrointestinal smooth muscle in several mammalian species, and all show similar affinity for .sup.125 I-porcine galanin (.about.0.5-1 nM). Nevertheless, recent in vitro and in vivo pharmacological studies in which fragments and analogues of galanin were used suggest the existence of multiple galanin receptor subtypes. For example, a galanin binding site in guinea pig stomach has been reported that exhibits high affinity for porcine galanin (3-29) (Gu, et al. 1995), which is inactive at CNS galanin receptors. The chimeric galanin analogue M15 (galantide) acts as antagonist at CNS galanin receptors (Bartfai et al., 1991) but as a full agonist in gastrointestinal smooth muscle (Gu et al., 1993). Similarly, the galanin-receptor ligand M40 acts as a weak agonist in RINm5F insulinoma cells and a full antagonist in brain (Bartfai et al, 1993a). The pharmacological profile of galanin receptors in RINm5F cells can be further distinguished from those in brain by the differential affinities of [D-Tyr.sup.2 ]- and [D-Phe.sup.2 ]-galanin analogues (Lagny-Pourmir et al., 1989). The chimeric galanin analogue M35 displaces .sup.125 I-galanin binding to RINm5F membranes in a biphasic manner, suggesting the presence of multiple galanin receptor subtypes, in this cell line (Gregersen et al., 1993).
Multiple galanin receptor subtypes may also co-exist within the CNS. Galanin receptors in the dorsal hippocampus exhibit high affinity for Gal (1-15) but not for Gal (1-29) (Hedlund et al., 1992), suggesting that endogenous proteolytic processing may release bioactive fragments of galanin to act at distinct receptors. The rat pituitary exhibits high-affinity binding for 12Sj-Bolton and Hunter (N-terminus) -labeled galanin (1-29) but not for [.sup.125 I] Tyr.sup.26 -porcine galanin (Wynick et al., 1993), suggesting that the pituitary galanin receptor is a C-terminus-preferring subtype. Spinal cord galanin binding sites, while similar to those in brain, show an affinity for the chimeric peptide antagonist M35 intermediate between the brain and smooth muscle (Bartfai et al., 1991), raising the possibility of further heterogeneity.
A galanin receptor cDNA was recently isolated by expression cloning from a human Bowes melanoma cell line (Habert-Ortoli et al., 1994). The pharmacological profile exhibited by this receptor is similar to that observed in brain and pancreas, and on that basis the receptor has been termed GALR1. The cloned human GALR1 receptor ("hGALR1") binds native human, porcine and rat galanin with .about.1 nM affinity (K.sub.i vs. .sup.125 I-galanin) and porcine galanin 1-16 at a slightly lower affinity (.about.5 nM). Porcine galanin 3-29 does not bind to the receptor. The GALR1 receptor appears to couple to inhibition of adenylate cyclase, with half-maximal inhibition of forskolin-stimulated cAMP production by 1 nM galanin, and maximal inhibition occurring at about 1 .mu.M.
Recently the rat homologue of GALRl ("rGALR1") was cloned from the RIN14B pancreatic cell line (Burgevin, et al., (1995), Parker et al., 1995. The pharmacologic data reported to date do not suggest substantial differences between the pharmacologic properties of the rat and human GALR1 receptors. Localization studies reveal GALR1 mRNA in rat hypothalamus, ventral hippocampus, brainstem, and spinal cord (Gustafson et al., 1996), regions consistent with roles for galanin in feeding, cognition, and pain transmission. However, GALR1 appears to be distinct from the pituitary and hippocampal receptor subtypes described above.
The indication of multiple galanin receptor subtypes within the brain underscores the importance of defining galanin receptor heterogeneity at the molecular level in order to develop specific therapeutic agents for CNS disorders. Pharmacological tools capable of distinguishing galanin receptor subtypes in tissue preparations are only beginning to appear. Several high-affinity peptide-based galanin antagonists have been developed and are proving useful in probing the functions of galanin receptors (see Bartfai et al., 1993), but their peptide character precludes practical use as therapeutic agents. In light of galanin's multiple neuroendocrine roles, therapeutic agents targeting a specific disorder must be selective for the appropriate receptor subtype to minimize side effects.
Accordingly, applicants have endeavored to clone the entire family of galanin receptors for use in target-based drug design programs. The identification of non-peptide agents acting selectively only at specific galanin receptors will be greatly facilitated by the cloning, expression, and characterization of the galanin receptor family.
Applicants have recently isolated by expression cloning from a rat hypothalamic cDNA library a novel galanin receptor, termed "GALR2," not described herein, which is distinguishable from GALR1 both by its unique sequence and distinct pharmacologic properties.
Applicants now report the isolation of a novel galanin receptor subtype, referred to herein as "GALR3," from a rat hypothalamic cDNA library. This discovery provides a novel approach, through the use of heterologous expression systems, to develop subtype selective, high-affinity non-peptide compounds that could serve as therapeutic agents for eating disorders, diabetes, pain, depression, ischemia, and Alzheimer's disease. Pathophysiological disorders proposed to be linked to galanin receptor activation include eating disorders, diabetes, pain, depression, ischemia, Alzheimer's disease and reproductive disorders. Accordingly, treatment of such disorders may be effected by the administration of GALR3 receptor-selective compounds. The presence of galanin binding sites in multiple CNS areas suggests that GALR3 receptors may also play a role in cognition, analgesia, sensory processing (olfactory, visual), processing of visceral information, motor coordination, modulation of dopaminergic activity, neuroendocrine function, sleep disorders, migraine, and anxiety.